A Simple CNI plugin based on Linux bridge and VXLAN for Kubernetes

bvcni is a Container Networking Interface (CNI) plugin, designed for Kubernetes clusters. It simplifies container networking management by utilizing a combination of Bridge and VXLAN

bvcni uses Linux-based bridges and VXLAN in its architecture, emulating the approach of Flannel's VXLAN mode

bvcni aims for simplicity, minimizing complex configurations.

• Bridge Networking: Connects containers within the same network bridge, enabling communication between them.
• VXLAN Networking: Establishes connections between containers using VXLAN virtual networks, ensuring network isolation and scalability.

• No other CNI should be installed. If another CNI was previously used, a reset is recommended.

• Docker Hub

Apply the following command when no other CNI is installed:

kubectl apply -f https://raw.githubusercontent.com/royroyee/bvcni/new_branch/bvcni.yaml

Note : Run the command when no other CNI is installed.

Prior to the CNI installation, the node will be in the NOT READY state. Once applied, a DaemonSet named bvcnid will be deployed to each node. The CNI configuration file will be generated in the /etc/cni/net.d directory, and the plugin binary files will be stored in /opt/cni/bin. A temporary file, tmp/reserved_ips, is also created for IP allocation.

{
  "cniVersion": "0.3.1",
  "name": "bvcni",
  "type": "bvcni",
  "podcidr": "10.244.1.0/24"
}

Afterwards, the node will transition to the READY state, and you will be able to use kubernetes.

NAME                                 READY   STATUS             RESTARTS          AGE
bvcnid-68n69                         1/1     Running               0              10s
bvcnid-hrqzh                         1/1     Running               0              10s
bvcnid-t86df                         1/1     Running               0              10s

NAME          STATUS   ROLES                  AGE     VERSION
k8s-master    Ready    control-plane,master   3d      v1.23.0
k8s-worker1   Ready    worker                 3d      v1.23.0
k8s-worker2   Ready    worker                 2d22h   v1.23.0

royroyee@k8s-master:~$ kubectl get pods -o wide
NAME      READY   STATUS    RESTARTS      AGE    IP           NODE          NOMINATED NODE   READINESS GATES
alpine1   1/1     Running      0         164m   10.244.2.2   k8s-worker2   <none>           <none>
alpine2   1/1     Running      0         85m    10.244.2.4   k8s-worker2   <none>           <none>

royroyee@k8s-master:~$ kubectl exec -it alpine1 -- ping 8.8.8.8
PING 8.8.8.8 (8.8.8.8): 56 data bytes
64 bytes from 8.8.8.8: seq=0 ttl=52 time=28.315 ms
64 bytes from 8.8.8.8: seq=1 ttl=52 time=28.116 ms
64 bytes from 8.8.8.8: seq=2 ttl=52 time=28.130 ms

�royroyee@k8s-master:~$ kubectl exec -it alpine1 -- ping 10.244.2.4
PING 10.244.2.4 (10.244.2.4): 56 data bytes
64 bytes from 10.244.2.4: seq=0 ttl=64 time=0.067 ms
64 bytes from 10.244.2.4: seq=1 ttl=64 time=0.097 ms

TBF

• There are occasional instances where the VXLAN ARP, FDB, and routing table do not update correctly during DaemonSet build

  • This can be resolved by deleting and re-running the daemonset (i.e., running it twice).

• Communication between nodes is possible, but there are issues with pod-to-pod communication across different nodes.

  • We are currently investigating the issue, and although there is a potential solution using eBPF, it involves complex aspects, so we are currently putting it on hold.

• There is an issue with the proper update of iptables settings.

  • If the iptables update has not occurred or if internet connectivity is not available within the pod, you must manually execute the following command (per current manual instructions):

You need to perform this on each node :

$ iptables -A FORWARD -s 10.244.0.0/16 -j ACCEPT

$ iptables -A FORWARD -d 10.244.0.0/16 -j ACCEPT 

$ iptables -t nat -A POSTROUTING -s 10.244.2.0/24 -j MASQUERADE

10.244.2.0/24 : node's PODCIDR

We plan to fix this issue as soon as possible.

This project is still under development and there are many areas that need improvement. Feedback and suggestions are always welcome through issues or pull requests

Younghwan Kim

MIT License